I have discovered (not that long ago) that most USB implementations suck. I only wish I had known (I should have known) sooner as I'm sure many of the DACs listed here could have benefited from a proper USB/SPDIF converter. It's frustrating because there are so many combinations to try out. And I can't but help think some of the DACs didn't get a fair shake - that they would have radically improved, especially if their built-in USB converters sucked.
And then there's the question of the effectiveness of USB/SPDIF converters. For example. I suspect the AP1 (at least without the PP PS supply) doesn't do anything. It sounded worse than the PWD2's built-in USB and Amos / Currawong said it sounded the same as his AGD M7's USB (which I felt sucked balls). The AP1 ain't cheap. It's even more expensive with the PP. At this price point, even if it did work, it's not viable for any DAC less than $2K. I would have liked to find out if the PP did anything, but the owner of the AP1 was already pissed that it didn't do jack and said "F that."
In terms of input differences with Gungnir - in order of goodness.
- USB Gen 2
- Coax from my PC (which sound better than the PWD2 USB implementation, which is already not bad)
- ...gap...
- USB Gen 1
I don't have my OR5 right now, it's on loan, but I will report back on it. The thing with the OR5 is that it's not worth it for the Gungnir. The OR5 as I have it configured is ~$1500. The Gungnir is $750. But cost aside, the Gungnir won't scale with the OR5 like the M7 or PWD2. The fact that the M7 or PWD2 will take i2s provides even another level of advantage. The USB Gen 2 option is $100, optimally scales with the Gungnir's inherent capabilities, and better matches the Gungnir's particularities. By this I mean the USB Gen 2 provides more clarity, really brings out the sweetness of the AKM chip, restores the soft sounds which were either dropped or compressed upward, adds more low level information. From memory, the OR5 only did a little bit to address the last three shortcomings. The OR5 mostly did what it does best: blacken the background, tighten up the sounds, making things more precise sounding, provide more clarity, etc. However, I'm not 100% sure at this point and will require more listening / testing. We are not talking about huge differences.